1. An Introduction to Remote Sensing NASA Remote Sensing Training Geo Latin America and Caribbean Water Cycle capacity Building Workshop Colombia, November 28-December 2, 2011 A RSET Applied Remote Sensing Education and Training A project of NASA Applied Sciences

2. What is Remote Sensing? Remote sensing is a method of obtaining information about the properties of an object without coming into physical contact with it.

3. Why use Remote Sensing to Study the Earth ? Provides visual Global information Complements ground-monitoring networks or provides information where there are no ground-based measurements Provides advance warning of impending environmental events and disasters.

4. How Do Satellites Make Measurements? Passive satellite sensors measure radiation reflected or emitted by the earth- atmosphere system - Radiance Radiance is converted to a geophysical parameter. Examples: Accumulated Rainfall Snow Cover Accumulated Rainfall Guatemala

5. Example of Remote Sensing Product Precipitation Radar from TRMM (Guatemala)

6. Types of satellite orbits Geostationary orbit Fixed’ above earth at ~36,000 km Frequent Measurements Limited Spatial Coverage Low Earth Orbit (LEO) - Polar (Aqua, Terra)) - Nonpolar (TRMM) Circular orbit constantly moving relative to the Earth at 160-2000 km Less Frequent measurements (< 2 times per day) Large (global) spatial Coverage

7. Low Earth Orbit (LEO): Orbiting at an altitude of 160-2,000 km. 7 Path of Satellite Low-Earth Orbits (LEO)

8. Ascending Orbit: Moving South to North when that portion of the orbit track crosses the equator. Low Earth Orbit: Orbiting at an altitude of 160-2,000 km. 8 Path of Satellite Low-Earth Orbits (LEO)

9. Descending Orbit: Moving North to South when that portion of the orbit track crosses the equator. 9 Low-Earth Orbits (LEO)

10. Aqua (“ascending” orbit) day time Terra (“descending”) Day time LEO Polar Orbiting

11. Near-polar, sun-synchronous, orbiting the Earth every 98.8 minutes, crossing the equator going north (daytime ascending) at 1:30 p.m. and going south (night time descending) at 1:30 a.m. The orbit track changes every day but will repeat on a 16 day cycle. This is true for Aqua, Terra, and TRMM. Aqua’s Orbit

12. Terra - Descending Aqua - Ascending When looking at an image of a piece of the orbit the two sensors will have opposite ‘tilts’. Daytime Orbits

13. TRMM (“ascending” orbit) LEO nonpolar Orbiting TRMM’s Low orbit allows its instruments to concentrate on the tropics. This image shows half the observations TRMM makes in a single day

14. Earth-Observing Satellites Sun-Synchronous: The satellite is always in the same relative position between the Earth and Sun. Equator-Crossing Time: The local apparent solar time when the satellite crosses the equator. Example: Terra has an equatorial crossing time of 10:30 am, and is called an “AM” or morning satellite. 14

15. 15 The orbit is defined as having a cross-track and an along-track direction. Direction of Satellite Motion Along-Track Direction Cross-Track Direction LEO Field-of-View (FOV)

16. 16 Direction of Satellite Motion Cross-Track Scanning “Cross-Track Scanning,” Scan mirror swings back and forth. Sensor observes pixels in sequence across track and along the direction of the satellite’s motion. Earth-Observing Satellites

17. 17 Direction of Satellite Motion Satellites in Low Earth Orbit have only an instantaneous Field- of-View (IFOV) LEO Field-of-View (FOV)

18. MODIS Orbit in 3D

19. Remote Sensing Resolutions Spatial resolution Temporal resolution Spectral resolution Radiometric resolution

20. Spatial Resolution  Spatial Resolution : A simple definition is the pixel size that satellite images cover.  Satellite images are organized in rows and column called raster imagery and each pixel has a certain spatial resolution. Nadir pixel size Off-nadir pixel size FOV IFOV Satelliteheight

21. cylindrical isotropic projection Native satellite view vs. map projection increasing pixel size bow-tie effect Flight direction Scan direction “BowTie” effect

22. 22 Spatial Resolution of NASA Satellite Data Products  High Spatial resolution 250x250m; 500x500 m; 1x1 km; 0.05x0.05 degrees Example: MODIS True Color Imagery (RGBs)  Moderate Spatial Resolution 0.25x0.25 degrees Example: TRMM precipitation products.  Low Spatial Resolution (Level 3) Primarily at 1 x 1 degree - derived from each data set’s native resolution product Example: AIRS surface air temperature

23. Example: NASA High Spatial Resolution Product 2x2 km resolution MODIS TERRA True Color Image over Southern California January 4 th, 2009 Source: NASA GSFC Rapidfire AERONET Subset for Fresno, CA

24. Example: NASA Moderate Spatial Resolution Product 0.25x0.25 degree TRMM Accumulated Rainfall over Guatemala

25. Example: NASA Low Spatial Resolution Product MERRA Monthly Precipitable Water 1.25 x 1.25 Degrees

26. Temporal Resolution of Remote Sensing Data The frequency at which data are obtained is determined by: -Type and height of orbit -Size of measurement swath

27. Temporal resolution of Polar Orbiting Satellites Example: Terra, Aqua Observations available only at the time of the satellite overpass. IR based observations available 2X a day (AIRS) Visible observations available 1X a day Polar regions may have several observations per day.

28. Temporal resolution of nonpolar satellites Example: TRMM Observations available only at the time of the satellite overpass. Observations available less than once a day Note: derived products available at 3-hourly

29. Remote Sensing – Resolutions Spectral resolution – The number and range of spectral bands. More bands = More information Radiometric resolution – The bandwidth of the individual spectral bands. Important for avoiding or taking advantage of “atmospheric windows”

30. Satellite data levels of processing and formats

31. 31 Levels of Data Processing Level 1Source Data: L1a are raw radiance counts and L1b are calibrated radiances (after applying calibration to L1a) Level 2Derived geophysical variables at the same resolution and location as Level 1 source data (after applying atmospheric correction, etc.) Level 2GLevel 2 binned data mapped on a uniform space- time grid (Example: OMI Tropospheric NO2 at 0.25x0.25 degree resolution) Level 3Geophysical variables mapped on a uniform space-time grid in derived spatial and/or temporal resolutions (Example: MODIS Temperature at1x1 degree resolution)

32. 32 Levels of Data Processing and Spatial Resolution Level 1 and Level 2 data products have the highest spatial and temporal resolution Level 3 products are derived products with equal or lower spatial and temporal resolution than Level 2 products. Available hourly, daily and for some products also monthly

33. Level 1 Products Level 2 Products Less Processing More Processing Orbital data Level 3 Products Global composites of level 2 products Used to produce Levels of Data Processing Orbital data

34. Level 1 Products Level 2 Products Orbital data Level 3 Products Global composites of level 2 products Used to produce More User Control Less User Control Harder to Use Easier to Use Levels of Data Processing Orbital data

35. Level 2 Example: Guatemala Precipitation Radar from TRMM (4x4 km)

36. Level 3 Example: TRMM Accumulated Rainfall 0.25x0.25 degree TRMM Accumulated Rainfall over Guatemala

37. Reprocessing: Applying a new algorithm to and entire data set. Forward Processing: Applying the current algorithm to newly acquired data. Important Terms for Level 2 and Level 3 Products

38. Data Versions For some NASA data products more than one version may be available Note: Giovanni products are the most recent data version publicly available For each level of processing versions of data are periodically released as retrieval algorithms or other sources of information improve, e.g. V001, V002, V003

39. Data Formats Text/ASCII pros: easy to read and examine the data right away (can read with used tools such as excel and GIS software) cons: large data files Binary – HDF, NetCDF pros: takes less space, more information (metadata,SDS) cons: need specific tools or code to read the data KML or KMZ (zipped KML) pros - easy 2D and 3D visualization of the data through free tools such as Google Earth. Data are very low volume 39

40. HDF Data Formats HDF is the standard format for most NASA data HDF files contain both data and metadata SDS - Each parameter within an HDF file is referred to as an SDS (Scientific Data Set) An SDS must be referenced precisely according to name when analyzing the data with your own computer code.

41. Accessing different data formats (Example: Giovanni Download Page) 41 GIF KMZ HDF NetCDF ASCII

42. Putting it all together: data file names Data product: 3- Hourly Rain Rate (mm/hr) Version 6 Date (June 30, 2011) Data format is HDF5 Level of Processing is L3 Version 6 Time (GMT hour: 21) Data Format (HDF5) 3B42.110630.21.6A.HDF.Z Level 3

43. Conclusions NASA satellite data formats are varied and the most appropriate depends on specific user needs Available data formats include, ASCII, HDF, NetCDF, and KMZ Satellite data vary in spatial resolution depending on instrument characteristics and the level of processing (L2, L2G, L3)